Generating Random Number Sequences Using Electroencephalography (EEG) Signals

Abstract

  Random numbers play an important role in many applications, such as gaming, generating encryption keys, modeling and simulating complex phenomena, software testing, and more. There are many algorithms, procedures, and phenomena that serve as sources for generating true random numbers or pseudo-random numbers. However, these traditional methods have some drawbacks, such as computational complexity and time consumption. Recent studies have demonstrated the effectiveness of using biometric markers as generators of random numbers, such as fingerprints, facial recognition, voice, iris patterns, and handwriting. However, these traditional vital signs also have weaknesses; they are inherited and unchangeable, and they are exposed to the public, which makes them susceptible to theft. Recent research has focused on studying other biomarkers as generators of random numbers, such as electroencephalography (EEG). Electroencephalography (EEG) provides a recording of the brain's electrical activity through electrodes placed in specific areas on the scalp. This recording is displayed as a series of waves that contain special and unique features, which can be extracted and processed to generate random numbers.

      In this research, we will present a scheme for generating random number sequences, each 512 bits long, based on the characteristics of brain signals related to imagined movement, using the EEG Motor Movement/Imagery Dataset. We examined the randomness of these sequences through statistical tests from the National Institute of Standards and Technology (NIST), and we studied the distinctiveness of these sequences through Hamming distance. Our findings indicate the potential use of these sequences as encryption keys in security applications.